Flaptekst

This book presents a study for speaker recognition rates for speech transmitted through Bluetooth channel as a degraded speech signals, while training phase is made by clean speech. This is based on the Mel-frequency cepstral coefficients (MFCC) for feature extraction from the speech. Different approaches for feature extractions was experimented; features from signal, feature from DCT, from signal plus DCT, features from DST, from signal plus DST, DWT of signal, and from signal plus DWT. The book introduces how Neural Network (NN) classification technique is used in the experiment as a classifier. In the training phase of experiments, the MFCCs and polynomial coefficients are estimated to form the feature vectors of the database. In the testing phase, similar features to that used in the training are extracted from the degraded or transmitted samples and used for matching. This book has investigated the process of speaker identification from speech transmitted over Bluetooth system. How the recognition rate for SNR above 0 dB is getting improved, and how the energy compaction property of the DCT allows robust features to be extracted from few samples.